'''

List flv and f4v medias, replace them with mp4 urls and update the content and projects.

'''

import csv

import logging

import re

from itertools import chain

import requests

from django.conf import settings

from django.contrib.sites.models import Site

from django.core.management.base import BaseCommand

from django.core import management

from lxml import etree

from ldt.ldt_utils import models

#this function replace bad suffixs and prefixs of some media URL

#by a new one, beginning with "http" and ending with ".mp4"

def to_https(source, vidpath, tomp4=1):

    '''

    to https

    '''

    if source[len(source)-3:len(source)] == 'MP4' or source[len(source)-3:len(source)] == 'mp4'\

        or not re.match(r".*\..{3}$", source):

        tomp4 = 0

    if tomp4 == 1:

        source = source[0:len(source)-3]+"mp4"

    if source.startswith("http://ftv.iri-research.org/"):

        return "https://media.iri.centrepompidou.fr/"+source[len("http://ftv.iri-research.org/"):]

    elif source[0:5] == "https":

        return source

    elif source[0:4] == "http" or source[0:4] == "sftp":

        return "https"+source[4:len(source)]

    elif source[0:7] == "/video/":

        return "https://media.iri.centrepompidou.fr"+source

    elif source[0:6] == "video/" or source[0:6] == "audio/":

        return "https://media.iri.centrepompidou.fr/"+source

    elif vidpath == 'rtmp://media.iri.centrepompidou.fr/ddc_player/video/regardssignes/' or \

        vidpath == 'rtmp://media.iri.centrepompidou.fr/ddc_player/mp4:video/regardssignes/':

        return "https://media.iri.centrepompidou.fr/video/regardssignes/"+source

    elif source[0:4] == "mp4:":

        if vidpath == 'rtmp://media.iri.centrepompidou.fr/ddc_player/':

            if re.match(r".*\..{3}$", source):

                return "https://media.iri.centrepompidou.fr/" + source[4:]

            return "https://media.iri.centrepompidou.fr/" + source[4:] + ".mp4"

    return "https://media.iri.centrepompidou.fr/video/ldtplatform/"+source

def number_of_contents(source):    #this counts the number of contents linked to a media

    '''

    number_of_contents

    '''

    return len(models.Content.objects.filter(media_obj_id=source.id))

def number_of_projects(source):

    '''

    number_of_projects

    '''

    if number_of_contents(source) > 0:

        return len(models.Project.objects.filter\

                (content=models.Content.objects.filter(media_obj_id=source.id)[0]))

    return 0

def construct_youtube_embed(source):

    '''

    construct youtube video oembed link

    '''

    if re.match(r".*feature=player_embedded.+", source) != None:

        return "http://www.youtube.com/oembed?url=http://youtube.com/watch?v="\

            + source[len(source)-11:] +"&format=json"

    return "http://www.youtube.com/oembed?url=" + source + "&format=json"

class Command(BaseCommand):

    '''

    Command class

    '''

    help = 'delete medias without contents, replace media\'s source by a new URL'

    def add_arguments(self, parser):

        '''

        add arguments

        '''

        parser.add_argument(

            '-f',

            '--force',

            dest='force',

            action='store_true'

        )

        parser.add_argument(

            '-p',

            '--path',

            dest='path',

            default=None

        )

    def construct_ldt_embed(self, ldtid):

        '''

        construct ldt embed

        '''

        return "http://{base_url}ldtplatform/ldt/embed/v3/config?json_url=" \

                   "http://{base_url}ldtplatform/ldt/cljson/id/{ldt_id}&" \

                   "player_id=player_project_{ldt_id}&" \

                   "ldt_id={ldt_id}".format(base_url=self.base_url, ldt_id=ldtid)

    def clean_media_project(self, element, newsrc=None):

        '''

        change media objects' videopath and source if necessary

        change project .ldt

        '''

        basesrc = element.src

        if self.force:

            element.videopath = ''

            element.save()

        if newsrc != None:

            if self.force:

                element.src = newsrc

                element.save()

            self.mycsvfile.writerow([

                "Media",

                basesrc,

                "Yes",

                "changing source/videopath",

                newsrc, "\'\'"

                ])

        if number_of_projects(element) == 0:

            self.mycsvfile.writerow([

                "Project",

                element.src,

                "Yes",

                "initializing object(no project)"

                ])

            if self.force:

                mycontentid = models.Content.objects.filter(media_obj_id=element.id)[0].iri_id

                try:

                    management.call_command('initfrontproject', mycontentid)

                except Exception:

                    self.mycsvfile.writerow([

                        "Project",

                        element.src,

                        "No",

                        "socket error"

                        ])

                    return

                self.stdout.write(" Initializing project", ending='')

            else:

                self.stdout.write(" Project has to be initialized ", ending='')

                return

        ldtproj = models.Project.objects.filter(

            content=models.Content.objects.filter(media_obj_id=element.id)[0]

        )

        for singleproject in ldtproj:

            root = etree.XML(singleproject.ldt.encode('utf-8'), self.parser)

            if root.xpath('medias/media') == []:

                self.stdout.write(" le .ldt ne contient pas de media", ending='')

                continue

            if root.xpath('medias/media')[0].get("video") != '':

                embedurl = self.construct_ldt_embed(singleproject.ldt_id)

                if self.force:

                    root.xpath('medias/media')[0].set("video", '')

                self.stdout.write(" changing videopath arg in .ldt ", ending='')

                self.mycsvfile.writerow([

                    "Project",

                    embedurl,

                    "Yes",

                    "changing .ldt /medias/media/video",

                    "\'\'"

                    ])

                singleproject.ldt = etree.tostring(root)

                singleproject.save()

                self.logger.info("%s DONE\n", embedurl)

        element.save()

    def __init__(self, *args, **kwargs):

        super(Command, self).__init__(*args, **kwargs)

        self.base_url = Site.objects.get_current().domain + settings.BASE_URL

        self.parser = etree.XMLParser(encoding='utf-8')

        self.logger = logging.getLogger(__name__)

        self.force = False

        self.mycsvfile = None

    def handle(self, *args, **options):

        '''

        handle

        '''

        path = options['path']

        self.force = options['force']

        if not path:

            path = 'mediaInformations.csv'

        try:

            csvfile = open(path, 'wb')

        except IOError:

            self.stdout.write('file can\'t be opened')

            self.logger.error('cant open file')

            return

        self.mycsvfile = csv.writer(csvfile)

        self.mycsvfile.writerow([

            "Object type",

            "which object",

            "Change ?",

            "What(if Y)/Why (if N)",

            "How"

            ])

        j = 0

        files1 = models.Media.objects.all() #this list contains every media

        for elem1 in files1:

            if number_of_contents(elem1) == 0:

                if self.force:

                    elem1.delete()  #if there is no content

                    #linked to the media, the media is removed for the database

                    self.stdout.write(" No content found, media has been removed")

                else:

                    self.stdout.write(" No content found, media will be removed")

                self.mycsvfile.writerow([

                    "Media",

                    elem1.src,

                    "Yes",

                    "deleting object (no content)"

                    ])

                j += 1

                continue

            if elem1.src.lower() == to_https(elem1.src, elem1.videopath).lower():

                self.clean_media_project(elem1)

            if re.match(r".*\.youtube\.com.*", elem1.src) != None \

                or re.match(r".*youtu\.be.+", elem1.src) != None:

                myembed = construct_youtube_embed(elem1.src)

                if requests.get(myembed).status_code == 404:

                    self.stdout.write("%s : Video doesn't exists"% elem1.src)

                    if number_of_projects(elem1) > 0:

                        ldtproj = models.Project.objects.get(

                            id=models.Content.objects.filter(

                                media_obj_id=elem1.id

                                )[0].front_project_id

                            ).ldt

                        root = etree.XML(ldtproj.encode('utf-8'), self.parser)

                        if root.xpath(

                                'annotations/content/ensemble/decoupage/elements/element'

                            ) == []:

                            if self.force:

                                elem1.delete()

                                self.stdout.write("video doesn't exist anymore : media deleted")

                            self.mycsvfile.writerow([

                                "Media/Content/Project",

                                elem1.src,

                                "Yes",

                                "deleting(Video doesn't exist anymore + empty projects)"

                                ])

                            j += 1

                else:

                    self.clean_media_project(elem1)

        if self.force:

            self.stdout.write("%s files deleted"%j)

        else:

            self.stdout.write("%s files to delete"%j)

        i = 0

        files = list(chain(

            models.Media.objects.filter(src__iregex=r".*.flv$"),

            models.Media.objects.filter(src__iregex=r".*.f4v$"),

            models.Media.objects.filter(src__iregex=r".*.m4v$"),

            models.Media.objects.filter(src__iregex=r".*.mp4$").exclude(src__iregex=r"^https://.*"),

            models.Media.objects.filter(src__iregex=r"^mp4:.*").exclude(src__iregex=r".*\..{3}$")

        ))

        for elem in files:

            self.stdout.write(" \n%s/%s files done"%(i+1, len(files)), ending='')

            i += 1

            if number_of_contents(elem) == 0:

                continue

            mysrc = elem.src

            newsource = to_https(elem.src, elem.videopath)

            try:

                res = requests.head(newsource, timeout=10).status_code

            except requests.ConnectionError:

                self.stdout.write(" connection error", ending='')

                self.logger.error("CONNECTION ERROR FOR %s", elem.title)

                try:

                    res = requests.head(elem, timeout=10).status_code

                except requests.ConnectionError:

                    self.mycsvfile.writerow([

                        "Media",

                        mysrc,

                        "No",

                        "connection error",

                        newsource

                        ])

                    continue

                except (requests.exceptions.MissingSchema, requests.exceptions.InvalidSchema):

                    self.mycsvfile.writerow([

                        "Media",

                        mysrc,

                        "No",

                        "missing schema on base source!",

                        newsource

                        ])

                    continue

                except requests.exceptions.Timeout:

                    self.mycsvfile.writerow([

                        "Media",

                        mysrc,

                        "No",

                        "TIMEOUT!",

                        newsource

                        ])

                    continue

                else:

                    self.mycsvfile.writerow([

                        "Media",

                        mysrc,

                        "No",

                        "use source link : website doesn't work with https",

                        newsource

                        ])

                    continue

            except (requests.exceptions.MissingSchema, requests.exceptions.InvalidSchema):

                self.stdout.write(" Missing schema !", ending='')

                self.logger.warning("MISSING SCHEMA FOR %s", elem.title)

                self.mycsvfile.writerow([

                    "Media",

                    mysrc,

                    "No",

                    "missing schema!",

                    newsource

                    ])

                continue

            except requests.exceptions.Timeout:

                self.stdout.write(" Timeout !", ending='')

                self.logger.warning("Timeout FOR %s", elem.title)

                self.mycsvfile.writerow([

                    "Media",

                    mysrc,

                    "No",

                    "TIMEOUT!",

                    newsource

                    ])

                continue

            if res > 400:

                try:

                    ressrc = requests.head(

                        to_https(elem.src, elem.videopath, 0),

                        timeout=10

                        ).status_code

                except (requests.exceptions.Timeout, requests.ConnectionError):

                    self.stdout.write(" can't access source/new files", ending='')

                    self.logger.warning("can't access %s", elem.title)

                    res = "connection error"

                    self.mycsvfile.writerow([

                        "Media",

                        mysrc,

                        "No",

                        "website doesn't exist anymore",

                        newsource

                        ])

                    continue

                if ressrc == 404:

                    self.stdout.write(" can't access source/new files", ending='')

                    self.logger.warning("can't access %s", elem.title)

                    self.mycsvfile.writerow([

                        "Media",

                        mysrc,

                        "No",

                        "can't access source/new files",

                        newsource

                        ])

                elif ressrc == 200:

                    self.stdout.write(

                        " file not transcoded yet :"

                        "keep source extension or wait transcoding to be done",

                        ending='')

                    self.logger.warning("%s not transcoded yet", elem.title)